Research Article
Tree-LSTM Guided Attention Pooling of DCNN for Semantic Sentence Modeling
@INPROCEEDINGS{10.1007/978-3-319-72823-0_6, author={Liu Chen and Guangping Zeng and Qingchuan Zhang and Xingyu Chen}, title={Tree-LSTM Guided Attention Pooling of DCNN for Semantic Sentence Modeling}, proceedings={5G for Future Wireless Networks. First International Conference, 5GWN 2017, Beijing, China, April 21-23, 2017, Proceedings}, proceedings_a={5GWN}, year={2018}, month={1}, keywords={Dynamic Convolutional Neural Network (CNN) Tree-Structured Long-Short Term Memory (Tree-LSTM) Attention Pooling Semantic Sentence Modeling}, doi={10.1007/978-3-319-72823-0_6} }- Liu Chen
Guangping Zeng
Qingchuan Zhang
Xingyu Chen
Year: 2018
Tree-LSTM Guided Attention Pooling of DCNN for Semantic Sentence Modeling
5GWN
Springer
DOI: 10.1007/978-3-319-72823-0_6
Abstract
The ability to explicitly represent sentences is central to natural language processing. Convolutional neural network (CNN), recurrent neural network and recursive neural networks are mainstream architectures. We introduce a novel structure to combine the strength of them for semantic modelling of sentences. Sentence representations are generated by Dynamic CNN (DCNN, a variant of CNN). At pooling stage, attention pooling is adopted to capture most significant information with the guide of Tree-LSTM (a variant of Recurrent NN) sentence representations. Comprehensive information is extracted by the pooling scheme and the combination of the convolutional layer and the tree long-short term memory. We evaluate the model on sentiment classification task. Experiment results show that utilization of the given structures and combination of Tree-LSTM and DCNN outperforms both Tree-LSTM and DCNN and achieves outstanding performance.